This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug target/52530] [4.8 regression] Many 64-bit execution failures on Solaris 10/11 with Sun as
- From: "hjl.tools at gmail dot com" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: Thu, 08 Mar 2012 15:06:32 +0000
- Subject: [Bug target/52530] [4.8 regression] Many 64-bit execution failures on Solaris 10/11 with Sun as
- Auto-submitted: auto-generated
- References: <bug-52530-4@http.gcc.gnu.org/bugzilla/>
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52530
--- Comment #2 from H.J. Lu <hjl.tools at gmail dot com> 2012-03-08 15:06:32 UTC ---
(In reply to comment #1)
> Ouch.
>
> Before the change, we always used "q" modifier for addresses, only in two
> special cases we emitted "l". This "q" modifier forced DImode address even for
> SImode operands, avoiding addr32 prefixes.
>
> The change - while correct - cancelled this optimization.
>
> So, please revert this patch ASAP.
>
Uros, can you revert it for me? Thanks.